Lifetime Risk of ESRD

Turin and colleagues1 present some fascinating estimates of both the lifetime and short-term risks of ESRD in this issue of JASN, exploiting the excellent epidemiologic database in Alberta, Canada. Based on a total population of nearly 2.9 million persons with nearly 26 million person-years of follow-up, they have been able to calculate actual lifetime event rates of ESRD (2.6% and 1.8% for men and women, respectively) rather than the synthetic estimates based on observed incidence rates applied to hypothetical populations.2 The advantage of this approach is not so much in increased accuracy of the estimate as in the ability to array the predictions by starting levels of kidney function. For example, for men 40 years of age, an estimated GFR (eGFR) of 45–59 ml/min per 1.73 m2 increases the lifetime risk of ESRD by almost 7-fold over men with an initial eGFR of 60–89 ml/min per 1.73 m2. The other useful addition of initial eGFR is the relative effect of moderately impaired renal function by age. For example, for young men with a starting eGFR of 45–59 ml/min per 1.73 m2, the short- and long-term probability of progressing to ESRD is substantial (3.4% at 10 years and 8.6% lifetime risk). However, for men who have attained the age of 70, an eGFR of 45–59 ml/min per 1.73 m2 confers less than one-third the short- and long-term risk.1 This should be exceedingly useful in counseling patients. Moderately decreased kidney function should not be cause for panic in elderly patients. However, a similar level of kidney function in young patients needs to be taken very seriously. One hopes the authors will continue to expand their analyses to include other risk variables such as diabetes, presence of albuminuria, BP levels, and other clinical indicators. In this study, the authors use the well-accepted definition of ESRD as the initiation of renal replacement therapy, either dialysis or transplantation. In both the United States and Canada, the rates of treated renal failure have stabilized in the last decade. However, the rate of treated renal failure in the United States is well over twice as large as in Canada (371 and 159 per million, respectively).3,4 Even among Caucasians, the rate of treatment in the United States is nearly twice that of Canada. Thus, the estimates of lifetime risk of ESRD generated by Kiberd and Clase2 based on the United States are still probably more relevant to the United States renal community, particularly in light of the nearly 3-fold greater incidence of ESRD among African Americans and the 50% increased risk of ESRD in Hispanic Americans. As the authors note, this is a treatment-defined event and will be subject to changes in practice patterns over time, as well as differences in acceptance criteria between countries or geographic areas. The authors themselves use a different definition of ESRD in their recent JAMA article on treatment rates for ESRD, in which case they use eGFR of ≤15 ml/min per 1.73 m2 to define ESRD.5 Based on the eGFR criteria, about one-half of the persons in Canada who arrive at ESRD actually receive renal replacement therapy. This varies greatly by age, from almost 90% of persons <45 years of age to about 7% of persons >75 years of age. Untreated ESRD in the elderly is the subject of considerable debate within the renal community6,7 and will not be addressed here except to note that “treated ESRD” is a somewhat slippery concept and will inevitably affect both short- and long-term risk estimates. Is lifetime risk a valuable tool for risk estimation or for patient counseling? The advantage of such estimations is to illustrate the fact that even rare events such as ESRD, which we typically express as number of cases per million population, add up over a lifetime. The downside of such estimates is that they can lead to unnecessary worry.8 Although ESRD does not carry the same fear factor as does cancer, a lifetime sentence to a thrice weekly dialysis regimen can be a scary thing. It sounds somewhat remote to say that one’s risk of ESRD is 160 in 1,000,000 in the next year. That sounds like a long shot lottery. No chance it is going to happen. However, a 2%–3% chance of ESRD is definitely within the realm of possibility. That almost sounds like a real thing. Others have noted the difficulty in interpreting lifetime estimates of disease incidence for patients and have recommended the short-term age conditional estimates of a 10- or even 20-year time horizon.9 The problem with lifetime risk is its relevance to the average person. Everyone has a 100% lifetime risk of dying of something. The public health advances of the early and mid-20th century largely eliminated infectious diseases as primary causes of mortality. This in turn led to remarkable increases in life expectancy. Now, people live long enough to die of something else, primarily cancer and cardiovascular diseases. This is evidenced by the very high lifetime risks associated with cancer (45% and 38% for men and women, respectively),10 coronary disease (49% and 31% for men and women, respectively),11 and diabetes (33% and 39% for men and women, respectively).12 Basically, if you live long enough, one of the major chronic diseases is going to get you. We can now add ESRD to that list of inevitable end-of-life problems. Lifetime risk also does not quite necessarily parallel short-term risk. For example, although African Americans have higher rates of invasive incident cancer, their lifetime risk is actually lower than that of Caucasians, because of their lower life expectancy.13 This is further evidence that the shorter-term estimates are of more clinical value to clinicians and patients. In the United States, it has been estimated that >20 million persons have CKD, which is >10% of the adult population.14 The vast majority of these persons will not progress to ESRD, either in terms of a treatment-defined or an eGFR-defined end point. Studies such as that presented by Turin et al. need to be expanded to include additional risk factors to help quantify the most important predictors of CKD progression so that scarce health resources can be directed where they will do the most good. Disclosures None. The opinions given here are those of the author and do not represent the opinion of the National Institute of Diabetes, and Digestive, and Kidney Diseases.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.